CN103995828B - A kind of cloud storage daily record data analysis method - Google Patents
A kind of cloud storage daily record data analysis method Download PDFInfo
- Publication number
- CN103995828B CN103995828B CN201410145688.2A CN201410145688A CN103995828B CN 103995828 B CN103995828 B CN 103995828B CN 201410145688 A CN201410145688 A CN 201410145688A CN 103995828 B CN103995828 B CN 103995828B
- Authority
- CN
- China
- Prior art keywords
- matrix
- row
- frequent
- daily record
- cloud storage
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/17—Details of further file system functions
- G06F16/174—Redundancy elimination performed by the file system
- G06F16/1748—De-duplication implemented within the file system, e.g. based on file segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/18—File system types
- G06F16/1805—Append-only file systems, e.g. using logs or journals to store data
- G06F16/1815—Journaling file systems
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention relates to a kind of cloud storage daily record data analysis method, step 1, the data to cloud storage daily record carry out preanalysis;Step 2, the cloud storage daily record data after preanalysis is calculated, obtain generating the frequent item set that relation maxim needs;Step 3, the frequent item set obtained according to step 2 generate the relation maxim of cloud storage daily record;The correlation rule present invention that step 4, output step 3 are obtained reduces the scale of the candidate matrix of generation by the abbreviation to frequent item set matrix, is effectively reduced the quantity of the candidate generated in successive iterations calculating process;In addition, in further improved technology scheme, invention calculates candidate matrix by customized matrix operation, whole calculating process is fairly simple, can reduce the operand in data analysis process, shortens the excavation time.
Description
Technical field
The invention belongs to data analysis technique field, more particularly to a kind of cloud storage daily record data analysis method can be used for
The data analysis of cloud storage system daily record.
Background technology
Cloud storage system can produce substantial amounts of journal file during operation.These journal files have recorded system
Keeper to the operational circumstances of system, when user is to the reception of access situation and system server, analysis request, the operation of system
The various raw informations such as mistake.Data analysis is carried out to system manager's Operation Log, can be with the operation of standardized administration person;To with
Family accesses situation daily record and carries out data analysis, it can be found that the behavioural habits of user, are conducive to inquiring about, analyze each user's operation,
Lifting user satisfaction;Data analysis is carried out to the daily record of cloud storage service device, network failure can be excluded with detecting system state,
Realize intrusion detection, additionally it is possible to find the mould that design defect, performance bottleneck and the needs of cloud storage system itself are distributed rationally
Block.
Because the daily record data amount produced by cloud storage system is very huge, thus how from substantial amounts of data, quickly
And valuable information is efficiently extracted out, the correlation between these information is found, as the analysis of cloud storage daily record data
In the primary problem for solving.Research currently for the data analysis of cloud storage system daily record is less.Relation maxim is data point
The important component of analysis process, by relation maxim it can be found that internal relation between mass data and valuable
System.Data analysis is carried out to cloud storage daily record, relation maxim is generated, it is possible to cloud storage journal file is effectively utilized.
At present, referring to shown in accompanying drawing 1, it mainly includes the flow of the data analysis of existing cloud storage daily record:Cloud is deposited
Storage journal file pretreatment, criterion generation and the criterion output to generating are analyzed using these steps.Wherein criterion life
Into this step, relate generally to find frequent item set and generation two aspects of relation maxim.Wherein find the method master of frequent item set
There are Apriori algorithm and the Apriori algorithm based on matrix.Apriori algorithm is that a kind of classical searching frequent item set is calculated
Method, the algorithm obtains latter using the alternative manner successively searched for using the result of calculation of previous item.Apriori algorithm quilt
Extensive research and improvement, it is the Apriori algorithm based on matrix that one of which is improved.This improvement is should by the thought of matrix
Use in the algorithm, analyzed database table is shown as the form of matrix.By this method can be by the scanning of database
Number of times is reduced to twice, shorten the data-analysis time, improves algorithm performance.
However, there is problems with the existing Apriori algorithm based on matrix:The amount of calculation of the algorithm first than larger,
When the data items included in analyzed database are more, the time that the algorithm expends can be exponentially increased, so right
When mass data is analyzed, more times can be expended;Secondly, the algorithm can produce excessive during being iterated
Candidate, stores these candidates meeting committed memory space, and increase amount of calculation when follow-up iterative calculation is carried out.
These shortcomings are unfavorable for from cloud storage daily record quickly extracting relation maxim, cause the data analysis of whole cloud storage daily record
Journey needs long time, inefficient, it is impossible to reflects the running status situation of cloud storage system in time, is unfavorable for entering system
Row optimization and performance boost.
The content of the invention
The technical problems to be solved by the invention are directed to above-mentioned prior art and provide a kind of cloud storage daily record data analysis
Method, the method can reduce the candidate data generated in amount of calculation and iterative process, can greatly improve the effect of data analysis
Really.
The present invention solve the technical scheme that is used of above-mentioned technical problem for:A kind of cloud storage daily record data analysis method,
It comprises the following steps:
Step 1, the data to cloud storage daily record carry out preanalysis, that is, delete the repeated data in daily record data, polishing day
Missing data in will data;
Step 2, the cloud storage daily record data after preanalysis is calculated, obtain generating the frequent episode that relation maxim needs
Collection;
Step 3, the frequent item set obtained according to step 2 generate the relation maxim of cloud storage daily record;
The correlation rule that step 4, output step 3 are obtained;
It is characterized in that:If constant k, k are the number of times of iterative calculation, it is 2 to make the initial value of k, in the step 2, is passed through
Following steps obtain generating the frequent item set that relation maxim needs:
Step 2a, using after preanalysis cloud storage daily record data generate the item collection Matrix C of candidate 11:
The item collection matrix of candidate 1The matrix is M row N column matrix, cijIt is the matrix
The element of the i-th row jth row, i and j is the item collection Matrix C of candidate 11Location index, wherein 1≤i≤M, 1≤j≤N,IjIt is j-th event recorded in cloud storage log database, Ij∈{I1,I2,…,IN, 1,2 ...
N represents the label of the event included in cloud storage log database, and N is the sum of event;TiIt is cloud storage daily record according in storehouse
I-th daily record, Ti∈{T1,T2,…TM, 1,2 ... M represent cloud storage daily record according in storehouse record daily record label, M is day
The sum of will;cijIt is a Boolean, 0 or 1 can only be taken, if i.e. i-th daily record T of cloud storage journal file recordiBag
Containing j-th event Ij, then cijTake 0, otherwise cijTake 1;
Step 2b, using given minimum support ScWith the item collection Matrix C of candidate 11, calculate frequent 1 item collection matrix L1, it is right
The matrix carries out abbreviation, obtains the frequent 1 item collection matrix L after abbreviation1', wherein minimum support ScN is multiplied by equal to constant x, often
The span of number x is 0~1;It is realized especially by following steps:
Step 2b-1, the calculating item collection Matrix C of candidate 11The row of each row and, successively by each row and with minimum support ScThan
Compared with, Ruo Lie and less than minimum support Sc, then the row are deleted, on the contrary then retain the row, obtain the first intermediary matrix;
Step 2b-2, calculate the first intermediary matrix per a line row and, if row and less than 2, by the row delete, otherwise protect
The row is stayed, the second intermediary matrix is obtained;
Step 2b-3, calculate the second intermediary matrix respectively row row and, successively by each row and with minimum support ScCompare,
If arranging and less than minimum support Sc, then the row are deleted, on the contrary then retain the row, generate new frequent 1 item collection matrix L1';
Step 2c, the number of times that k is iterative calculation is set, it is 2 to make the initial value of k, and the value of (k-1) is tried to achieve using the value of k, it is determined that
Go out to calculate candidate's k item collection Matrix CskWhen frequent (k-1) the item collection matrix L that needsk-1', and according to frequent (k-1) the item collection matrix
Lk-1' obtain candidate's k item collection Matrix Csk:
Step 2d, using minimum support ScWith candidate's k item collection Matrix Csk, calculate frequent k item collections matrix Lk, and to this frequently
Numerous k item collections matrix LkAbbreviation is carried out, the frequent k item collections matrix L after abbreviation is obtainedk', comprise the following steps:
Step 2d-1, the calculating item collection Matrix C of candidate 22In each row row and, successively by each column arrange and with minimum support Sc
Compare, if row are less than minimum support Sc, then the row are deleted, on the contrary then retain the row, obtain the 3rd intermediary matrix;
Step 2d-2, the value that k+1 is obtained according to the value of iterative calculation number of times k, calculate row of the 3rd intermediary matrix per a line
With, if going and less than k+1, corresponding row being deleted and is gone, otherwise retain the row, obtain the 4th intermediary matrix;
Step 2d-3, again calculate the 4th intermediary matrix respectively row row and, successively by each row and with minimum support Sc
Compare, Ruo Lie and less than minimum support Sc, then the row are deleted, on the contrary then retain the row, the frequent k item collections square after generation abbreviation
Battle array Lk';
Step 2e, judge the frequent k item collections matrix L after abbreviationk' whether it is empty matrix, if the matrix is empty matrix, tie
Beam is calculated, and otherwise makes k from increasing 1, repeat step 2c to step 2d.
As improvement, candidate k item collection Matrix Cs in the step 2ckIt is calculated by following manner:
IfWherein u, v difference table are frequent (k-1) item collection matrix Lk-1' position rope
Draw, and according to frequent (k-1) the item collection matrix Lk-1' obtain candidate's k item collection Matrix Csk:
Wherein " ∧ " represents AND operator.
Improve again, the step 3 obtains relation maxim as follows:
Step 3a, given min confidence Sz, min confidence SzSpan be 0~1;
Step 3b, for abbreviation after frequent k item collections matrix Lk', by Lk' each row constitute one-column matrix l, l a ∈
Lk', produce all nonvoid subset r of one-column matrix l;
Step 3c, for each nonvoid subset r, ifWherein SrIt is the counting of nonvoid subset r, SlIt is single-row square
Battle array l row and, then obtain relation maxim
Compared with prior art, the advantage of the invention is that:The present invention is reduced by the abbreviation to frequent item set matrix
The scale of the candidate matrix of generation, is effectively reduced the quantity of the candidate generated in successive iterations calculating process;
In addition, in further improved technology scheme, invention calculates candidate matrix by customized matrix operation, whole to calculate
Process is fairly simple, can reduce the operand in data analysis process, shortens the excavation time.
Brief description of the drawings
Fig. 1 is the data analysing method flow chart of cloud storage daily record in the prior art;
Fig. 2 is that step 2 obtains the flow chart of frequent item set in the embodiment of the present invention;
Fig. 3 is using generation frequent item set method in the generation frequent item set method in the embodiment of the present invention and existing method
Effect contrast figure.
Specific embodiment
The present invention is described in further detail below in conjunction with accompanying drawing embodiment.
Cloud storage daily record data analysis method as shown in Figure 2, it comprises the following steps:
Step 1, the data to cloud storage daily record carry out preanalysis, that is, delete the repeated data in daily record data, polishing day
Missing data in will data;
Step 2, the cloud storage daily record data after preanalysis is calculated, obtain generating the frequent episode that relation maxim needs
Collection, in this step, if constant k, k are the number of times of iterative calculation, it is 2 to make the initial value of k, this step again through the following steps that
Obtain generating the frequent item set that relation maxim needs:
Step 2a, using after preanalysis cloud storage daily record data generate the item collection Matrix C of candidate 11:
The item collection matrix of candidate 1The matrix is M row N column matrix, cijIt is the matrix
The element of the i-th row jth row, i and j is the item collection Matrix C of candidate 11Location index, wherein 1≤i≤M, 1≤j≤N,IjIt is j-th event recorded in cloud storage log database, Ij∈{I1,I2,…,IN, 1,2 ...
N represents the label of the event included in cloud storage log database, and N is the sum of event;TiIt is cloud storage daily record according in storehouse
I-th daily record, Ti∈{T1,T2,…TM, 1,2 ... M represent cloud storage daily record according in storehouse record daily record label, M is day
The sum of will;cijIt is a Boolean, 0 or 1 can only be taken, if i.e. i-th daily record T of cloud storage journal file recordiBag
Containing j-th event Ij, then cijTake 0, otherwise cijTake 1;
Step 2b, using given minimum support ScWith the item collection Matrix C of candidate 11, calculate frequent 1 item collection matrix L1, it is right
The matrix carries out abbreviation, obtains the frequent 1 item collection matrix L after abbreviation1', step 2b is to frequent 1 item collection matrix L1Abbreviation, is changed
Frequent 1 item collection matrix L after letter1', wherein minimum support ScN is multiplied by equal to constant x, the span of constant x is 0~1;Its
Realized especially by following steps:
Step 2b-1, the calculating item collection Matrix C of candidate 11The row of each row and, successively by each row and with minimum support ScThan
Compared with, Ruo Lie and less than minimum support Sc, then the row are deleted, on the contrary then retain the row, obtain the first intermediary matrix;It is wherein minimum
Support ScN is multiplied by equal to constant x, the span of constant x is 0~1;
Step 2b-2, calculate the first intermediary matrix per a line row and, if row and less than 2, by the row delete, otherwise protect
The row is stayed, the second intermediary matrix is obtained;
Step 2b-3, calculate the second intermediary matrix respectively row row and, successively by each row and with minimum support ScCompare,
If arranging and less than minimum support Sc, then the row are deleted, on the contrary then retain the row, generate new frequent 1 item collection matrix L1';
Step 2c, the number of times that k is iterative calculation is set, it is 2 to make the initial value of k, and the value of (k-1) is tried to achieve using the value of k, it is determined that
Go out to calculate candidate's k item collection Matrix CskWhen frequent (k-1) the item collection matrix L that needsk-1', and according to frequent (k-1) the item collection matrix
Lk-1' obtain candidate's k item collection Matrix Csk, candidate k item collection Matrix Cs in step 2c herekIt is calculated by following manner:
IfWherein u, v difference table are frequent (k-1) item collection matrix Lk-1' position rope
Draw, and according to frequent (k-1) the item collection matrix Lk-1' obtain candidate's k item collection Matrix Csk:
Wherein " ∧ " represents AND operator;
Step 2d, using minimum support ScWith candidate's k item collection Matrix Csk, calculate frequent k item collections matrix Lk, and to this frequently
Numerous k item collections matrix LkAbbreviation is carried out, the frequent k item collections matrix L after abbreviation is obtainedk', here to frequent k item collections matrix in step 2d
LkAbbreviation is carried out, the frequent k item collections matrix L after abbreviation is obtainedk', comprise the following steps:
Step 2d-1, the calculating item collection Matrix C of candidate 22In each row row and, successively by each column arrange and with minimum support Sc
Compare, if row are less than minimum support Sc, then the row are deleted, on the contrary then retain the row, obtain the 3rd intermediary matrix;
Step 2d-2, the value that k+1 is obtained according to the value of iterative calculation number of times k, calculate row of the 3rd intermediary matrix per a line
With, if going and less than k+1, corresponding row being deleted and is gone, otherwise retain the row, obtain the 4th intermediary matrix;
Step 2d-3, again calculate the 4th intermediary matrix respectively row row and, successively by each row and with minimum support Sc
Compare, Ruo Lie and less than minimum support Sc, then the row are deleted, on the contrary then retain the row, the frequent k item collections square after generation abbreviation
Battle array Lk';
Step 2e, judge the frequent k item collections matrix L after abbreviationk' whether it is empty matrix, if the matrix is empty matrix, tie
Beam is calculated, and otherwise makes k from increasing 1, repeat step 2c to step 2d;
Frequent item set after step 3, the abbreviation obtained according to step 2 generates the relation maxim of cloud storage daily record, specific bag
Include:
Step 3a, given min confidence Sz;
Step 3a, given min confidence Sz, min confidence SzSpan be 0~1;
Step 3b, for abbreviation after frequent k item collections matrix Lk', by Lk' each row constitute one-column matrix l, l a ∈
Lk', produce all nonvoid subset r of one-column matrix l;
Step 3c, for each nonvoid subset r, ifWherein SrIt is the counting of nonvoid subset r, SlIt is single-row square
Battle array l row and, then obtain relation maxim
The correlation rule that step 4, output step 3 are obtained.
Effect of the present invention can be further illustrated by following emulation:
1. simulated conditions
Emulation of the invention is the hardware environment and MATLAB in dominant frequency 2.5GHz intel (R) Core (TM) i5CPU
Carried out under the software environment of R2009b, Window 7.
2. emulation content and interpretation of result
Using frequent item set method is generated in the generation frequent item set method in the present invention and existing method, as a result such as Fig. 3
It is shown.
It can be seen from figure 3 that the run time of two methods is all reduced with the increase of minimum support, but fortune of the invention
The row time is significantly lower than existing method, i.e., the present invention when frequent item set is generated efficiency apparently higher than existing method.This be due to
Present invention amount of calculation when candidate is generated it is small, it is necessary to time it is few, and abbreviation has been carried out to frequent item set, make generation
Candidate quantity is significantly reduced.
Simulation result shows:The present invention generates candidate as a result of customized matrix operation, reduces life
Into the amount of calculation of candidate, and the quantity that abbreviation reduces the candidate of generation is carried out by frequent item set, accelerate cloud
The efficiency of storing daily record data analysis.
Claims (3)
1. a kind of cloud storage daily record data analysis method, it comprises the following steps:
Step 1, the data to cloud storage daily record carry out preanalysis, that is, delete the repeated data in daily record data, polishing daily record number
Missing data in;
Step 2, the cloud storage daily record data after preanalysis is calculated, obtain generating the frequent item set that relation maxim needs;
Step 3, the frequent item set obtained according to step 2 generate the relation maxim of cloud storage daily record;
The correlation rule that step 4, output step 3 are obtained;
It is characterized in that:If constant k, k are the number of times of iterative calculation, it is 2 to make the initial value of k, in the step 2, by following
Step obtains generating the frequent item set that relation maxim needs:
Step 2a, using after preanalysis cloud storage daily record data generate the item collection Matrix C of candidate 11:
The item collection matrix of candidate 1The matrix is M row N column matrix, cijIt is the i-th row of the matrix
The element of jth row, i and j is the item collection Matrix C of candidate 11Location index, wherein 1≤i≤M, 1≤j≤N,IjIt is j-th event recorded in cloud storage log database, Ij∈{I1,I2,…,IN, 1,2 ...
N represents the label of the event included in cloud storage log database, and N is the sum of event;TiIt is cloud storage daily record according in storehouse
I-th daily record, Ti∈{T1,T2,…TM, 1,2 ... M represent cloud storage daily record according in storehouse record daily record label, M is day
The sum of will;cijIt is a Boolean, 0 or 1 can only be taken, if i.e. i-th daily record T of cloud storage journal file recordiBag
Containing j-th event Ij, then cijTake 0, otherwise cijTake 1;
Step 2b, using given minimum support ScWith the item collection Matrix C of candidate 11, calculate frequent 1 item collection matrix L1, to the square
Battle array carries out abbreviation, obtains the frequent 1 item collection matrix L after abbreviation1', wherein minimum support ScN is multiplied by equal to constant x, constant x's
Span is 0~1;It is realized especially by following steps:
Step 2b-1, the calculating item collection Matrix C of candidate 11The row of each row and, successively by each row and with minimum support ScCompare, if
Arrange and less than minimum support Sc, then the row are deleted, on the contrary then retain the row, obtain the first intermediary matrix;
Step 2b-2, calculate the first intermediary matrix per a line row and, if row and less than 2, by the row delete, otherwise retain should
OK, the second intermediary matrix is obtained;
Step 2b-3, calculate the second intermediary matrix respectively row row and, successively by each row and with minimum support ScCompare, if row
With less than minimum support Sc, then the row are deleted, on the contrary then retain the row, generate new frequent 1 item collection matrix L1';
Step 2c, the value that (k-1) is tried to achieve using the value of k, determine calculating candidate's k item collection Matrix CskWhen frequent (k-1) item that needs
Collection matrix Lk-1', and according to frequent (k-1) the item collection matrix Lk-1' obtain candidate's k item collection Matrix Csk;
Step 2d, using minimum support ScWith candidate's k item collection Matrix Csk, calculate frequent k item collections matrix Lk, and it is frequent k to this
Collection matrix LkAbbreviation is carried out, the frequent k item collections matrix L after abbreviation is obtainedk', comprise the following steps:
Step 2d-1, the calculating item collection Matrix C of candidate 22In each row row and, successively by each column arrange and with minimum support ScCompare,
If row are less than minimum support Sc, then the row are deleted, on the contrary then retain the row, obtain the 3rd intermediary matrix;
Step 2d-2, the value that k+1 is obtained according to the value of iterative calculation number of times k, calculate row of the 3rd intermediary matrix per a line and, if
Go and less than k+1, then corresponding row is deleted and gone, otherwise retain the row, obtain the 4th intermediary matrix;
Step 2d-3, again calculate the 4th intermediary matrix respectively row row and, successively by each row and with minimum support ScCompare,
If arranging and less than minimum support Sc, then the row are deleted, on the contrary then retain the row, the frequent k item collections matrix L after generation abbreviationk';
Step 2e, judge the frequent k item collections matrix L after abbreviationk' whether it is empty matrix, if the matrix is empty matrix, terminate meter
Calculate, otherwise make k from increasing 1, repeat step 2c to step 2d.
2. cloud storage daily record data analysis method according to claim 1, it is characterised in that:Candidate k in the step 2c
Item collection Matrix CkIt is calculated by following manner:
IfWherein u, v difference table are frequent (k-1) item collection matrix Lk-1' location index, and root
According to frequent (k-1) the item collection matrix Lk-1' obtain candidate's k item collection Matrix Csk:
Wherein " ∧ " represents AND operator.
3. cloud storage daily record data analysis method according to claim 1, it is characterised in that:The step 3 is by as follows
Step obtains relation maxim:
Step 3a, given min confidence Sz, min confidence SzSpan be 0~1;
Step 3b, for abbreviation after frequent k item collections matrix Lk', by Lk' each row constitute one-column matrix l, a l ∈ Lk',
Produce all nonvoid subset r of one-column matrix l;
Step 3c, for each nonvoid subset r, ifWherein SrIt is the counting of nonvoid subset r, SlIt is one-column matrix l
Row and, then obtain relation maxim
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410145688.2A CN103995828B (en) | 2014-04-11 | 2014-04-11 | A kind of cloud storage daily record data analysis method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410145688.2A CN103995828B (en) | 2014-04-11 | 2014-04-11 | A kind of cloud storage daily record data analysis method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103995828A CN103995828A (en) | 2014-08-20 |
CN103995828B true CN103995828B (en) | 2017-06-13 |
Family
ID=51309993
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410145688.2A Expired - Fee Related CN103995828B (en) | 2014-04-11 | 2014-04-11 | A kind of cloud storage daily record data analysis method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103995828B (en) |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104881467B (en) * | 2015-05-26 | 2018-08-31 | 上海交通大学 | Data correlation analysis based on frequent item set and pre-reading method |
CN106991141B (en) * | 2017-03-21 | 2020-12-11 | 北京邮电大学 | Association rule mining method based on deep pruning strategy |
CN108536841A (en) * | 2018-04-13 | 2018-09-14 | 国家计算机网络与信息安全管理中心 | Intelligence learning interrogation model based on service-user custom |
CN108830655A (en) * | 2018-06-19 | 2018-11-16 | 郑州云海信息技术有限公司 | A kind of user's operation Relation acquisition method and relevant apparatus |
CN109947728B (en) * | 2019-03-11 | 2023-09-05 | 中国银联股份有限公司 | Log file processing method and device |
CN111125183B (en) * | 2019-11-07 | 2023-06-23 | 北京科技大学 | Tuple measurement method and system based on CFI-Apriori algorithm in fog environment |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7370033B1 (en) * | 2002-05-17 | 2008-05-06 | Oracle International Corporation | Method for extracting association rules from transactions in a database |
CN103020256A (en) * | 2012-12-21 | 2013-04-03 | 电子科技大学 | Association rule mining method of large-scale data |
CN103279570A (en) * | 2013-06-19 | 2013-09-04 | 广西教育学院 | Text database oriented matrix weighting negative pattern mining method |
-
2014
- 2014-04-11 CN CN201410145688.2A patent/CN103995828B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7370033B1 (en) * | 2002-05-17 | 2008-05-06 | Oracle International Corporation | Method for extracting association rules from transactions in a database |
CN103020256A (en) * | 2012-12-21 | 2013-04-03 | 电子科技大学 | Association rule mining method of large-scale data |
CN103279570A (en) * | 2013-06-19 | 2013-09-04 | 广西教育学院 | Text database oriented matrix weighting negative pattern mining method |
Non-Patent Citations (3)
Title |
---|
关联规则中频繁项集高效挖掘的研究;张云涛等;《计算机工程与应用》;20111231;第139-141页 * |
基于向量和矩阵的频繁项集挖掘算法研究;张敏等;《计算机工程与设计》;20130331;第34卷(第3期);第939-943页 * |
基于项目集格及位图索引的频繁项目集发现算法;陈富赞等;《系统工程理论与实践》;20080229(第2期);第26-34页 * |
Also Published As
Publication number | Publication date |
---|---|
CN103995828A (en) | 2014-08-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Mehmood et al. | Implementing big data lake for heterogeneous data sources | |
CN103995828B (en) | A kind of cloud storage daily record data analysis method | |
Yang et al. | A system architecture for manufacturing process analysis based on big data and process mining techniques | |
CN105912666B (en) | A kind of mixed structure data high-performance storage of facing cloud platform, querying method | |
CN104112026B (en) | A kind of short message text sorting technique and system | |
CN111343161B (en) | Abnormal information processing node analysis method, abnormal information processing node analysis device, abnormal information processing node analysis medium and electronic equipment | |
CN106960006A (en) | Measuring similarity system and its measure between a kind of different tracks | |
CN113254630B (en) | Domain knowledge map recommendation method for global comprehensive observation results | |
CN105389341A (en) | Text clustering and analysis method for repeating caller work orders of customer service calls | |
CN117971606B (en) | Log management system and method based on elastic search | |
CN116881430B (en) | Industrial chain identification method and device, electronic equipment and readable storage medium | |
CN110121729A (en) | For identification with the system and method for the signal for including in characterize data stream | |
CN109308290A (en) | A kind of efficient data cleaning conversion method based on CIM | |
CN113779261A (en) | Knowledge graph quality evaluation method and device, computer equipment and storage medium | |
CN113505117A (en) | Data quality evaluation method, device, equipment and medium based on data indexes | |
CN109213793A (en) | A kind of stream data processing method and system | |
CN110442614B (en) | Metadata searching method and device, electronic equipment and storage medium | |
Li et al. | Automatic classification algorithm for multisearch data association rules in wireless networks | |
Hassan et al. | Sampling technique selection framework for knowledge discovery | |
Mathai et al. | An efficient approach for item set mining using both utility and frequency based methods | |
CN107562909A (en) | A kind of big data analysis system and its analysis method for merging search and calculating | |
JP6201053B2 (en) | Feature data management system and feature data management method | |
Qing-dao-er-ji et al. | An improved HotSpot algorithm and its application to sandstorm data in Inner Mongolia | |
de Sousa et al. | Integrated detection and localization of concept drifts in process mining with batch and stream trace clustering support | |
CN117539920B (en) | Data query method and system based on real estate transaction multidimensional data |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170613 |